The first trading day of 2026 delivered a stark message about the evolving artificial intelligence trade: hardware is hot, software is not. While Nvidia climbed more than 1% and Micron rocketed 7% higher, cloud computing stalwarts Salesforce and CrowdStrike declined meaningfully, extending a divergence that has been building for months.

The Great AI Divide

Investors have increasingly bifurcated the technology sector into two camps: companies that directly benefit from AI infrastructure spending, and everyone else. The numbers are telling.

In 2025, the Themes Cloud Computing ETF (CLOD) gained just 7%—barely a third of the Nasdaq-100's 21% return. Within that ETF, the divergence was even more pronounced. While Snowflake surged 44.6% and CrowdStrike gained 39.4%, Salesforce dropped nearly 20% despite beating earnings estimates in four consecutive quarters.

"We're seeing a fundamental re-rating of software companies. Investors are questioning whether these businesses can maintain growth and margins in an AI-disrupted world."

— Brad Sills, Analyst, Bank of America Securities

Why Hardware Is Winning

The semiconductor sector benefits from clear, measurable demand for AI infrastructure. According to Goldman Sachs, hyperscaler capital expenditure could reach $527 billion in 2026, with a substantial portion flowing directly to chip companies.

Nvidia, AMD, and Micron are selling picks and shovels in a gold rush. Every major cloud provider, enterprise IT department, and nation-state AI initiative needs their chips. The demand is real, the orders are placed, and the revenue is visible.

Key semiconductor trends driving the rally:

  • HBM memory shortage: High-bandwidth memory chips used in AI systems are sold out through 2026, benefiting Micron and SK Hynix
  • Data center buildout: AI training clusters require thousands of GPUs and specialized memory
  • Next-generation chips: Nvidia's Blackwell and AMD's MI300 architectures command premium pricing
  • China demand: Chinese tech companies have ordered over 2 million Nvidia H200 chips for 2026 delivery

Software's Identity Crisis

For software companies, the AI picture is murkier. While enterprise software vendors talk enthusiastically about AI features, investors worry about two distinct threats: commoditization and disruption.

The commoditization concern is straightforward. If AI assistants can help developers write code faster, does enterprise software become cheaper to build? If large language models can automate customer service, do companies need as many software seats? These questions hang over valuations.

The disruption concern is more existential. What happens when AI agents can configure software, manage workflows, and automate tasks that currently require human oversight? Some investors believe AI could compress the software value chain, eliminating layers of applications that exist primarily to manage complexity.

Salesforce's Agentforce Pivot

Salesforce's decline is particularly instructive. The company has pushed aggressively into AI with its Agentforce platform, which promises to automate customer service and sales tasks using AI agents. Yet investors remain skeptical, sending shares down sharply even as the company delivers solid quarterly results.

"The narrative surrounding Salesforce is shifting from fear to execution. While critics previously labeled the company an 'AI loser,' Agentforce is showing real traction. But the market wants to see monetization, not just demos."

— Keith Weiss, Analyst, Morgan Stanley

Management argues that Agentforce will eventually drive higher per-seat pricing and improved customer retention. But in the near term, the transition creates uncertainty about growth rates and competitive positioning.

Cybersecurity's Mixed Picture

Cybersecurity stocks present a more nuanced case. CrowdStrike recovered from its catastrophic outage in July 2024 to post a 39.4% gain in 2025, yet shares remain well below their all-time highs. Palo Alto Networks and Fortinet have traded sideways despite cybersecurity spending remaining robust.

Investors appear to be pricing in both the opportunity and the risk that AI presents. On one hand, AI-powered security tools could command premium pricing. On the other, AI could commoditize basic security functions, squeezing margins.

Gartner's Optimistic Outlook

Not everyone is bearish on software. Research firm Gartner forecasts that enterprise software spending will grow 15.2% in 2026 to $1.43 trillion, making it the fastest-growing segment of the $6 trillion IT market.

The bulls argue that software stocks are due for mean reversion. If AI truly drives productivity gains, enterprises will spend more on software, not less. And the companies that successfully integrate AI into their platforms could emerge with even wider competitive moats.

Investment Implications

For investors navigating this divergence, several themes emerge:

  • Hardware remains the cleaner trade: Demand visibility for semiconductors exceeds that for software
  • Selectivity matters in software: Companies with clear AI monetization strategies may outperform
  • Valuation compression creates opportunities: High-quality software at depressed multiples could offer attractive entry points
  • Watch for convergence: Eventually, AI benefits should flow to software companies that successfully adapt

The AI trade is evolving from a simple "buy anything with AI in the name" thesis to a more nuanced evaluation of which companies actually benefit from the technology. In 2026, the gap between hardware winners and software laggards may narrow—or it may widen further as investors get clarity on who the real AI beneficiaries are.

For now, the first trading day of the year reinforced a clear message: when it comes to AI, making the chips beats making the software.